Scripts2026年3月29日·1 分钟阅读

Diffusers — Universal Video & Image Generation Hub

Hugging Face's diffusion model library. Run CogVideoX, AnimateDiff, Stable Video Diffusion, and 50+ video/image models with a unified API. 33,200+ stars.

TO
TokRepo精选 · Community
快速使用

先拿来用，再决定要不要深挖

这里应该同时让用户和 Agent 知道第一步该复制什么、安装什么、落到哪里。

pip install diffusers torch
import torch
from diffusers import CogVideoXPipeline
from diffusers.utils import export_to_video

pipe = CogVideoXPipeline.from_pretrained("THUDM/CogVideoX-2b", torch_dtype=torch.float16)
pipe.to("cuda")
video = pipe("A golden retriever playing in snow").frames[0]
export_to_video(video, "output.mp4")
介绍

Hugging Face Diffusers is the universal hub for running diffusion models — including 50+ video generation models (CogVideoX, AnimateDiff, Stable Video Diffusion, Wan Video, LTX-Video, and more) through a single unified API. Pre-trained models download automatically from Hugging Face Hub. 33,200+ GitHub stars.

Best for: AI researchers, developers building video generation features, anyone wanting to run state-of-the-art video models locally Works with: Python, PyTorch, NVIDIA GPUs (CUDA)

Part of the Video AI Toolkit collection.

Supported Video Models

Model Type Quality
CogVideoX Text-to-video High
AnimateDiff Image-to-video High
Stable Video Diffusion Image-to-video High
Wan Video Text-to-video High
LTX-Video Text+audio-to-video High
ModelScope Text-to-video Good

Image Generation Models

Also supports 100+ image models: Stable Diffusion, SDXL, Flux, Kandinsky, PixArt, and more.

FAQ

Q: What is Diffusers? A: Hugging Face's Python library for running diffusion models with a unified API. Supports 50+ video generation models and 100+ image models. 33,200+ stars.

Q: Is Diffusers free? A: Yes. Apache 2.0 licensed. Models are freely downloadable from Hugging Face Hub. Requires GPU for practical use.

🙏

来源与感谢

Created by Hugging Face. Licensed under Apache 2.0. diffusers — ⭐ 33,200+ Docs: huggingface.co/docs/diffusers

相关资产

MoneyPrinterTurbo — One-Click Short Video Generator

AI-powered end-to-end short video pipeline: topic in, finished video out. Auto-generates script, materials, subtitles, and background music. 53,000+ stars.

TokRepo精选

VideoCaptioner — AI Subtitle Pipeline

LLM-powered video subtitle tool: Whisper transcription + AI correction + 99-language translation + styled subtitle export. 13,800+ stars.

TokRepo精选

Open-Sora — Open-Source Text-to-Video Generation

Open-source alternative to Sora by HPC-AI Tech. Generate videos from text prompts with an 11B parameter model. Apache 2.0 licensed. 28,800+ stars.

TokRepo精选
◈首页🏆排行榜👤我的